A three-gene random forest model for diagnosing idiopathic pulmonary fibrosis based on circadian rhythm-related genes in lung tissue

ABSTRACT

Background

The disorder of circadian rhythm could be a key factor mediating fibrotic lung disease Therefore, our study aims to determine the diagnostic value of circadian rhythm-related genes (CRRGs) in IPF.

Methods

We retrieved the data on CRRGs from previous studies and the GSE150910 dataset. The participants from the GSE150910 dataset were divided into training and internal validation sets. Next, we used several various bioinformatics methods and machine learning algorithms to screen genes. Next, we identified SEMA5A, COL7A1, and TUBB3, which were included in the random forest (RF) diagnostic model. Finally, external validation was conducted on data retrieved from the GSE184316 datasets.

Results

The results revealed that the RF diagnostic model could diagnose patients with IPF in the internal validation set with the area under the ROC curve (AUC) value of 0.905 and in the external validation with the AUC value of 0.767. Furthermore, real-time quantitative PCR and western blotting results revealed a significant decrease in SEMA5A (p < 0.05) expression level and an increase in COL7A1 and TUBB3 expression levels in TGF-β1-treated normal human lung fibroblasts.

Conclusion

We constructed an RF diagnostic model based on SEMA5A, COL7A1, and TUBB3 expression in lung tissue for diagnosing patients with IPF.

KEYWORDS:

• Idiopathic pulmonary fibrosis
• circadian rhythm
• diagnosis
• machine learning
• random forest

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

J He designed the study, provided financial support, and conducted the experiments. J Hu and H Liu developed the algorithm and conducted the experiments. J He wrote the original draft. H Liu helped to collect the clinical samples and edited the original manuscript. All authors reviewed the manuscript and approved the final version of the manuscript.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17476348.2024.2311262

Additional information

Funding

This paper was funded by Sichuan Medical Research Project Foundation (S21054) and Priming Scientific Research Foundation for the Introduced Talents of The First Affiliated Hospital of Chengdu Medical College (CYFY-GQ59).